GB2528301A

GB2528301A - Biomolecule separation using apatite

Info

Publication number: GB2528301A
Application number: GB1412692.4A
Authority: GB
Inventors: Alastair Hawkins
Original assignee: University of Newcastle, The; Newcastle University of Upon Tyne
Current assignee: University of Newcastle, The; Newcastle University of Upon Tyne
Priority date: 2014-07-17
Filing date: 2014-07-17
Publication date: 2016-01-20
Also published as: WO2016009221A1; GB201412692D0

Abstract

The invention relates to a method of separating a biomolecule of interest from one or more contaminant species comprising contacting a buffer-equilibrated apatite matrix with a sample containing the biomolecule of interest and the one or more contaminant species, washing the matrix with an apatite matrix wash solution and collecting fractions of the apatite matrix flow-through which contain the biomolecule of interest and wherein the biomolecule of interest is separated from the one or more contaminant species; wherein upon contact with and transit through the apatite matrix the biomolecule and the one or more contaminant species do not become adsorbed to the matrix and wherein no desorption is performed; and wherein the biomolecule of interest is separated from the one or more contaminant species following differential rate of transit of the biomolecule and the one or more contaminant species through the apatite matrix. Also claimed is a method for producing a purified biomolecule solution from a contaminated biomolecule solution comprising using: i) ion exchange matrix chromatography, and ii) an apatite matrix separation step. Also claimed are associated kits containing an apatite matrix and an ion exchange matrix. Preferably, the apatite matrix is hydroxyapatite.

Description

BIOMOLECULE SEPARATION USING APATITE

FIELD OF THE INVENTION

[001] This invention relates to methods for the purification of biomolecules such as polypeptides using apatite matrices, in particular hydroxyapatite matrices. The methods involve the separation of polypeptides not by conventional techniques using apatite matrices involving an adsorption step and a desorption step. Rather, the methods involve a technique which exploits a differendal rate of ilansit of a biomolecule of interest and one or more contaminant species through the apatite matrix. The invention also relates to the use of apatite, and particularly hydroxyapatite, materials in such methods and kits for use in such methods. The invention additionally relates to such methods and uses for the purification of secreted nuclease enzymes, and in particular the deoxyribonuclease enzyme NucB derived from Bacillus iu'henijorrnis, and enzymes related thereto.

BACKGROUND TO THE INVENTION

[002] Existing techniques for the separation of molecules, in particular biomolecijlcs. and in particular proteins, are many and varied. For instance, in size exclusion chromatography (such as gel fihration), molecules are separated on the basis of their size, A solution containing a mixture of molecules, including the molecule of interest, is passed over a colunm of porous beads. Molecules are distributed outside and inside the pores on the basis of their size. Large molecules which cannot fit into the pores pass through the column first. Molecules which are retained on the beads will pass through the column in a manner dependent upon their size, with smaller molecules eluting later.

[003] Ion exchange chromatography exploits the fact that molecules, typically proteins, carry charge and therefore can interact with a matrix carrying an opposite charge. In this technique, proteins can be separated on the basis that the protein of interest may have a charge profile different from other molecules in the mixture, and therefore can he adsorbed onto the separation matrix, and desorhed therefrom, differently and in a manner dependent upon charge. This technique can offer very good separation of proteins from complex mixtures [004] Many other separation techniques are available and are commonly used, including mixed-mode separadon techniques. In contrast to techniques like gel filtration and ioll exchange, mixed-mode techniques exploit multiple aild more complex modes of interaction between the native molecule and the separation matrix.

Apatite minerals, and in particular hydroxyapatite, are one such group of materials that operate on a mixed-mode basis and will be described in more detail below.

[005] Separation materials and techniques all have their limitations, not least since biomolecules of interest very often have biophysicallphysiochemical properties which are similar to those of contaminant biomolecules. Therefore, a biomolecule of interest and a contaminating hiornolecule(s) may behave in a similar manner with respect to their interaction with a separation matrix and eg. may co-elute together with the hiornolecule of interest from the matrix. For this reason it is often necessary to employ multiple purification techniques in series to achieve appropriate separation leading to purification. Nevertheless, even with the use of multiple techniques it is often very difficult to separate a given biomolecule of interest from a complex mixture of molecules.

[006] The present disclosure describes a technique for the separation of hiomolecules, in particular proteins, and in particular a certain class of proteins, which involves apatite material as the separation matrix, and iii particular hydroxyapatite.

Prior to this disclosure it has proved difficult to purify such proteins to homogeneity due to contaminant species whose complete separation therefrom could not be achieved by known techniques. Moreover and in addition, it has surprisingly been found that apatite material can be employed to separate a biomolecule of interest, pat-ticulat-ly a protein, from contaminant proteins by what is believed to be a novel mechanism of action.

[007] l-lydroxyapatite (Ca10(P04)6(0H)2). fluorapatite (Ca10(P04)6(F)2) and chlorapatite (Ca10(P04)6(Cl)2) comprise a group of crystal! inc phosphate mineral materials known collectively as apatite. They are characterised as having pairs of positively charged crystal calcium ions (calcium loci or C-sites) and triplets of crystal phosphate groups (phosphate loci or P-sites). Uvdroxyapatite has clusters of six negatively charged oxygen atoms associated with each triplet phosphate P-site.

Hydroxyapatite and tluoroapatite in particular have received attention as reagents for use in chromatographic separation of molecules and particularly biomolecules. For instance, calcium, phosphate and hydroxyl groups are arranged throughout the crystalline structure of hydroxyapatite and present a differentially charged surface suitable for binding certain bioinolecules. However, the mechanism of interaction of these apatite materia!s with hiomolecules is comp!ex. and as a result they have tended to be less well used than other materials in chromatographic processes.

[00$] Gorhunoff I)] conducted a systematic study of desorption of proteins from hydroxyapatite matrices as a function of (a) the isoelectric point of the protein. (h) the ionic nature of the e!uent used in the desorption process and (c) structural differences between related proteins. It was found that certain basic proteins could be grouped together based on their elution from the matrix by similar (moderate) niolarities of P04. F, CU, SCN and C104. Certain acidic proteins could be grouped together based on their elution from the matrix by equal (again moderate) mnolarities of POt and F, hut which proteins did not elute with Ca2, and usua!ly did not e!ute with Cl-.

Certain neutral proteins could be grouped together based on their clution from the matrix by P043-, F and C[, hut which did not elute with Ca2 or SCN. This study a!so revea!ed that individual specific po!ar groups were generally not eritica! to adsorption or desorption and that moderate variation in protein structure did not generally influence the interaction with hydroxyapatite. although adsorption was significantly disrupted if the tertiary structure of the protein was loosened or reduced to the state of a random coil. The study essentially established that desorption from hydroxyapatite is influenced by the net charge (isoelectric point) of the native globulai-protein. There was found to be no symmetry as between acidic aM basic proteins and their adsorption/desorption patterns, suggesting that acidic and basic residues interact with hydroxyapatite in a non-identical manner.

OO9J A further study by Gorhunoff 2j examined desorption from hydroxyapatite, using a variety of eluants, of certain proteins and derivatives thereof The derivatives were designed to alter the charge profile of the proteins. These studies again confirmed that elution patterns of native globular proteins are intluenced by their isoelectric points. It was also found that the adsorption of basic proteins requires the presence of a high density of positively charged groups and that adsorption of proteus equilibrated with phosphate buffer is enhanced by a decrease in negative charge of the protein, e.g. as a result of loss of cai-boxyl coups. On the other hard. clusters of cai-boxyl groups, as would be found in acidic proteins, were found to sbengthen the binding of proteins to hydroxyapatite and this binding was disrupted by calcium ions.

hdicating that calcium ions effect desorption at the level of earboxyl groups.

OO1OJ More specifically, amino groups were found to he crucial for the adsorption of basic proteins to hydroxyapatite hut not for the adsorption of acidic proteins, since the blocking of the amino groups abolished or greatly reduced binding of basic proteins but had little effect on the binding of acidic proteins. The mechanism of the interaction of amino groups with hydroxyapatite, however, was shown to be general in that the positive charge is essential, not the chemical nature of the group. With regard to the interaction of cai-hoxyl groups with hydroxyapatite, blocking such groups enhanced the adsorption of all proteins, hut replacement with a different negatively charged group (SOH) reduced binding of all proteins, and reduced the binding of acidic proteins to a much gi-eater degree. This suggested that the negatively charged carhoxyl groups are iepelled iioii-specifically fiom hydroxyapatite, a mechanism which predomiiiates iii basic proteins, hut that in acidic proteins carhoxyl groups contribute specifically to binding.

[00111 Thus, the interaction between proteins and hydroxyapatite was suggested to be complex, involving both general electrostatic interactions and specific binding effects, and not by a mechanism consistent with conventional ion exchange [21. The interaction between carhoxyl groups of acidic proteins and hydroxyapatite was suggested to involve binding between carhoxyl groups and C-sites, which are a feature of hydroxyapatite crystals [2]. However, such binding, to form coordination complexes at calcium loci, may require specific geometry such as might he a feature of a protein with a high surface density of carhoxyl groups. The interaction between amino groups of acidic proteins and hydroxyapatite was suggested to involve general electrostatic interactions betwcen amino groups and P-sitcs, also a feature of hydroxyapatite crystals [3]; phosphate loci also acting to repel protein carboxyl groups. A general mechanism of interaction with hydroxyapatite was proposed [3].

Positively charged protein amino residues thus interact with P-sites by phosphoryl cation exchange, whilst negatively charged protein carboxyl residues interact with calcium ions of C-sites by metal chelation. However, whilst the charge and its distribution on the interacting protein are important, the crystalline structure of hydroxyapatite, and the arrangement of phosphate groups, oxygen atoms and calcium atoms in the crystal structure. imposes a three-dimensional stereochemical aspect to the interaction, such that the tertiary structure of the interacting protein is also important L3, 4]. The complex nature of the interactions between amino and carhoxyl groups of proteins and hydroxyapatite consequently defines separation by hydroxyapatite as a mixed-mode or multiinodal process.

[0012] Subsequent studies established that phosphoryl groups present on molecules were capable also of interacting with calcium loci on hydroxyapatite crystals leading to a strong binding interaction and that other hiornolecules such as DNA were amenable to punfication using hydroxyapatitc 141.

[0013] These studies thus confirmed the utility of apatite, and iii particular hydroxyapatite, in the separation of hiomolecules, particularly proteins, and defined certain generally-applicable principles in this regard. Importantly. however, such prior uses of hydroxyapatite for the separation of proteins and other biomolecules has

S

been predicated on a mechanism requiring adsorption to hydroxyapatite and subsequent desoiption therefrom.

[0014] In a typical ad.sorption/desorption protocol for the separation of a protein of interest, the sample is loaded in low ionic strength phosphate buffer (e.g. 1-10mM sodium or potassium phosphate). Desorption is performed by the use of an elution buffer of much increased concentration and by using a concentration gradient (e.g..O- 400mM phosphate buffer). However, since basic, acidic and neutral proteins behave differently with respect to their adsorption to hydroxyapatite, and since amino and carboxyl groups also interact differently, the choice of elution buflèr and gradient concenfltion range will vary depending upon the physico-ehernical properties of the protein of interest. Nevertheless, phosphate buffer can be of broad applicability since with respect to a protein having basic properties, phosphate's ionic character may act to disrupt the interaction between amine coups and crystal phosphates; and with respect to a protein having acidic properties, phosphate's affinity for calcium may act to disrupt the coordination complexes between carhoxyl groups and crystal calcium.

Chloride buffer may also be used for elution purposes.

[0015] 1-lydroxyapatite may find use as a chromatographic reagent either in its native form or in metal-derivatized forms and in particular as a calcium-derivatized form.

Calciuni-derivatization leads to the replacement of P-sites with additional C-sites.

This consequently reduces the phorsphoryl cation exchange interaction with amino goups and increases the metal chelation interaction between calcium ions and carboxyl oups. As a result, the specificity of the material for the biomoleeule to be captured can he altered. More recently, improvements have been made in the manufacture of hydroxyapatite materials for protein separation purposes. For example, the agglomeration of hydroxyapatite into particles and subsequent sintcring at high temperatures leads to the formation of stable, porous, eel-arnie niicrospheres which have improved flow properties and capacities. Such ceramic hydroxyapatite materials are available commercially. (-I

[0016] Hydroxyapatite, however, is often considered disadvantageous for biomolecule separation for a variety of reasons. Because of the complex chemistries involved in adsorption and desorption of biomolecules, such as proteins, the multi-modal mechanisms ate far less straightforward than conventional ion exchange mechanisms, which can he considered to he simple reversible interactions and therefore much easier to perform As a result, hydroxyapatite chromatography is much more involved. Related to this, and as described above, differently charged proteins interact with hydroxyapatite in different ways As such, desorption may require different salts or ionic species used at varying concentrations, again adding to the involved nature of the technique. In addition, the presence in subsequent ethates of high concentrations of salts or ionic species, such as phosphate or chloride, which are required for desorption of certain proteins can interfere with steps downstream of separation, thus requiring significant dilution of the eluate or requiring buffer exchanges adding additional steps and complexity into protocols. As such, hydroxyapatite is not generally considered to be a routine choice of reagent for protein purification purposes and is often avoided for these reasons.

[0017] However, although generally perceived as being disadvantageous for protein separation by adsorption/desorption methods. hydroxyapatite has nevertheless been used in certain instances.

[0018] EP0256836 describes the separation of various species of Tissue Plasminogen Activator (tPA) by contachng a hydroxyapatite mathx with a crude tPA preparation, thus adsorbing various tPA species to the matrix. Subsequent desorption of tPA species of varying specific molecular weights was achieved using eluates of defined p1-I andlor salt concentrations [0019] In publication W0201 2/040216 Gagnon described adsorption/desorption methods using hydroxyapatite for the separation of desired hiomolecule products from contaminants. The methods were specifically devised to aid the dissociation of aggregates between the protein of interest and contaminants complexed with the protein of interest. The specific methods are illustrated by the dissociation of an IgG molecule of interest floin contaminating DNA coinpiexed with the IgG. The coinplexed mixture was loaded onto a calciurn-derivatized hydroxyapatite matrix and washed with specific decoinpiexant wash buffers to aid in the dissociation of DNA from the IgU. In all cases the matrix was then eluted with a linear gradient to I M NaCI. DNA was retained on the column and was collected in a subsequent cleaning step. It was postulated that the decomplexant wash initially weakens the association between DNA and IgU and, due to its strong affinity for the hydroxyapatite calcium, DNA preferentially associates with the matrix after the deconiplexant wash allowing subsequent elution of the lgG whilst retaining DNA on the matrix.

[0020] In publication W02009/092010, Gagnon described various other adsorption/desorption methods using hydroxyapatite for the separation of desired biomoleeule products from contaminants. The described method involved contacting a hydroxyapatite matrix with an impure prepat-ation and eluting the matrix in the presence of an ionic species which is a sulphate. borate, monocarboxylic organic acid salt or monocarhoxylic zwitterion.

[0021] Unlike the presently-claimed uses and methods, however, all of these prior uses of hydroxyapatite for protein separation are predicated on a step of adsorption to the hydroxyapatite matrix of the protein of interest and then subsequent desorption, or alternatively adsorption and retention on the hydroxyapatite matrix of all contaminant species.

[0022] Thus there remains a need for alternative biomolecule purification methods involving apatite materials, in particular hydroxyapatite, which does not suffer from the drawbacks of desorption steps requiring high salt concentrations which can interfere with subsequent sample processing. Additionally, there remains a need for alternative hiomolecule purification methods involving apatite which can provide for the separation of a hiomolecule of interest and at least one contaminant in situations where both species do not become adsorhed to the apatite matrix and therefore would co-elute from the matrix. There remains a particular need for alternative hiomolecule purification methods where a biomolecule of interest cannot be separated from one or more contaminant species using oniy an ion exchange rnaix and wherein the biornolecule of interest and one or more contaminant species cannot be separated by adsorpdon/desorption methods involving another separation matrix including an apatite matrix. There also remains a need for alternative hiomolecule purification methods where a hiomolecule of interest cannot he separated from one or more contaminant species using only a size exclusion matrix (such as gel filtration) and wherein the hiomolecule of interest and one or more contaminant species cannot he separated by adsorption/desorption methods involving another separation matrix including an apatite matrix.

[0023] The purification of biomolecules, and more specifically proteins and polypeptides, can pose a variety of challenges for other reasons. For example, high level expression of proteins in microorganisms such as bacteria can lead to the formation of inclusion bodies composed of aggregated and sometimes incorrectly folded protein which can hamper subsequent purification. For certain proteins of interest, heterologous expression in higher eukaryotic cells, such as insect or mammalian cells, may lead to expressed proteins which possess a more similar post-transladonal modification profile compared to the native protein, but may be produced at low levels. As well as potential basic expression problems, proteins of interest may often he expressed together with a range of other hiomolecules, and therefore purifying the protein of interest, once expressed. may pose further and different challenges.

[0024] Secreted microbial nuclease enzymes are one class of proteins the purification of which may pose challenges once expressed. Being naturally secreted, it is necessary to purify such proteins from conditioned culture media as a starting material. However, this can lead to consequential problems. Nuclease enzymes are secreted naturally by microbes to disrupt a constituent known as hiofilm. Biofilm is a complex polymeric matrix composed of a variety of hiomolecules including carbohydrates, poly-and oligo-saccharides and a species refelTed to as extracellular DNA or eDNA. These and other constituents may additionally be secreted by the inhabiting microbes. Biofilm forms a protective capsule defining a local environment which the microbes inhabit. Recently it has been discovered that microbes appear to secrete nuclease enzymes for the purpose of modifying the eDNA component of biofilm in a manner which may allow microbes actively to alter the snctural architecture of hiofilm [5]-In this way, microbes may impose a degree of control over their natural habitat. As a consequence, homologous or heterologous expression of naturally-secreted microbial nucleases, particularly if they are expressed in microbial host cells, may require conditions which also promote the expression of a range of other hiomolecules. Such other hiomolecules, which may he naturally advantageous for the organism as a component of hiofilm, may he undesirable as a co-expressed contaminant since the nuclease of interest must then be separated therefrom. This is in addition to any other co-expressing contaminating biomolecules which may also be secreted into the culture medium. Consistent with this, it has been found that with respect to one such secreted microbial nuclease enzyme. the deoxyribonuclease enzyme NucB derived from Bacillus licheniformis, when subjected to conditions designed to promote maximal heterologous expression the conditioned culture medium produced comprises a complex mixture of hiomolecules. This has consequences in terms of obtaining consistency (of both yield and purity) of a final purified preparation of the NucB enzyme.

[0025] Thus there remains a need in the art for additional techniques for the purification of secreted nuclease enzymes, more particularly for secreted deoxyribonuclease enzymes, and more particularly for the deoxyribonuclease enzyme NucB derived froui Bacillus lichenijbrniis, so as to achieve consistently high yields and purity.

SUMMARY OF THE INVENTION

[0026] The present invention is based on the surprising finding that a biomolecule of interest may be separated from one or more contaminant species via a differential Uansit rate separation method of purification using an apatite maix. By using methods involving the differential nnsit rate separadon techniques described herein, preparations of certain biomolecules of interest may be achieved having high and consistent yields and being of a high percentage purity.

[0027] In such methods, the hiomolecule of interest is one which will not adsorb to an apatite matrix. Moreover the "one or more contaminant species". as described in the methods defined herein, will also not adsorb to the apatite matrix, and would theiefore otheiwise co-elute with the hiomolecule of inteiest in a hulk flow-thi-ough from the apatite matrix, making it difficult to separate from the biomolccule of interest using conventional adsorptionldesorption methods. The one or more contaminant species contrasts with other more general "contaminating species" which may be capable of adsorption to certain separation matrices. Thus, although the biomolecule of interest may be separated from certain contaminating species using conventional separation matrices and methods, the methods described herein may be used in order to finafly separate the hiomolecule of interest from the one or more contaminant species that will also not adsorb to an apatite matrix.

[0028] The invention provides a method of separating a hiomolecule of interest from one or more contaminant species comprising contacting a buffer-equilibrated apatitc matrix with a sample containing the biomolecule of interest and the one or more contaminant species. washing the matrix with an apatite matrix wash solution and collecting fractions of the apatite maftix flow-through which contain the biomolecule of interest and wherein the biomolecule of interest is separated from the one or more contaninant species; wherein upon contact with and transit through the apatite matrix the hiomolecule and the one or more contaminant species do not become adsorbed to the matrix and wherein no desorption step is performed; and wherein the hiomolecule of interest is separated from the one or more contaminant species following differential rates of transit of the hiornolecule and the one or more contaminant species through the apatite matrix.

[0029] Typically, as well as being unable to adsorb to an apatite matrix the biomolecule of interest and the one or more contaminant species, as described in the methods defined herein, may also be difficult or unable (for practical purposes) to adsorb to an ion exchange matrix. As such, the bioniolecule of interest will be one which is difficult to sepai-ate, or cannot (for pracdcal purposes) be sepa-ated, from the one or more contaminant species using only an ion exchange matrix, and would therefore co-elute with the hiomolecule of interest in a bulk flow-through or in the same elution fraction from an ion exchange matrix. Thus, the hiomolecule of interest may he a hiomolecule that cannot he separated from the one or more contaminant species using an ion exchange matrix.

0030] Typically, as well as being unable to adsorb to an apatite matrix the biomolecule of interest may also be difficult to separate. or may be unable (for practical purposes) to be separated, from the one or more contaminant species only by a size exclusion mauix. A size exclusion matrix includes a gel filntion matrix, As such, the biomolecule of interest and the one or more contaminant species would co-elute with the hiomolecule of interest in a bulk flow-through or in the same fraction from a size exclusion matrix. Thus, the hiomolecule of interest maybe a hiomolecule that cannot he separated from the one or more contaminant species using a size exclusion matrix including a gel filtration matrix.

[0031] A yet further advantage of the methods described herein is that because there are no adsorption and desorption steps, it is not necessary to change the buffer compositions during the various steps of the procedure. Thus, in the methods described herein the apatite matrix equilibration solution, the sample containing the biomolecule of interest and the one or more contaminant species, and the apatite matrix wash solution may all be comprised of a buffered solution having the same composition.

[0032] Thus one, more or all of the ahove-mentioned buffered solntions may comprise potassium phosphate pH7.2 at a concentration of between 10-75mM, preferably 50mM potassium phosphate pH7.2.

[0033] One, more or all of the above-mentioned buffered solutions may further comprise DTT at a concentration of between 0.5-5mM, preferably 1mM DTT.

[0034] One, more or all of the above-mentioned buffered solutions may comprise 50mM potassium phosphate pU72 and 1mM DTT.

[0035] In any of the methods defined herein, the biomoleeule of interest may be a polypeptide. In any of the methods defined herein, the biomolecule of interest may be a recombinant polypeptide.

[0036] In any of the methods defined hei-ein, the polypeptide may have a neutral charge at pH 6.8-7.2.

[0037] In any of the methods defined herein, the polypeptide may have a molecular weight of approx. 12 kDa.

[0038] In any of the methods defined herein, the polypeptide may he a microbial deoxyrihonuclease enzyme. The polypeptide may he a secreted lfliclohial deoxyrihonuclease enZyme.

[0039] In any of the methods defined herein, the polypeptide may he a microbial deoxyrihonuclease enzyme selected from the list of bacterial deoxyrihonuclease, gram positive bacterial deoxyribonuelease, class Bacillus bacterial deoxyribonuelease, Bacillus lichemformes bacterial deoxyribonuclease, Bacillus lichenejonnis strain El- 34-6 bacterial deoxyribonuelease. microbial NueB deoxyribonuclease. or any enzymatically-aetive truncate, fragment. homolog, ortholog, paralog, analog or derivative thereof [0040] In any of the methods defined herein, the polypeptide may he a microbial deoxyrihonuclease enzyme which is Bacillus licheniforinis strain EI-34-6 NucB deoxyrihonuclease defined by SEQ ID NO: I, or any enzymatically-active truncate, fragment, homolog, ortholog, paralog, analog or derivative thereof [0041] As noted previously, the biornolecule of interest is one which will not adsorb to an apatite matrix. Moreover, the one or more contaminant species, as described in the methods defined herein, will also not adsorb to the apatite matrix and would therefore co-elute with the hiomolecule of interest in a hulk flow-through from the apatite matrix, making it difficult to separate from the hiomolecule of interest. Thus, although the hiomolecule of interest may he separated from certain contaminating species using conventional separation matrices and methods, the methods defined herein may he used in order to finally separate the hiomolecule of interest from one or more specific contaminant species (as defined in the methods described herein) that will also not adsorb to an apatite matrix. Thus, the defined one or more contaminating species that does not adsorb to an apatite matrix and which is/are desired to be separated from the biomolecijle of interest using the methods defined herein may be a biomolecule or biomolecules selected from the list of peptides, polypeptides, proteins, carbohydrates, saccharides. oligosaccharides. polysaccharides, lipids, oligonucleotides, polynucleotides. nucleic acids, or any combination thereof.

[0042] For all practical purposes. other than the constituents of the buffer the collected apatite matrix flow-through fractions contain only the hiomolecule of interest and are essentially and substantially free of the one or more contaminant species and any other contaminating species. Thus. in the methods described herein all of the one or more contaminant species which are present in the sample may be absent from the collected apatite matrix flow-through fractions.

[0043] In methods described herein, in the collected apatite matrix flow-through fractions which contain the hiomolecule of interest, the hiomolecule of interest may comprise 90% or more of all hiomolecule species in the fraction, or may comprise 95% or more of all biomolecule species in the fraction.

[0044] Certain methods described herein may further comprise combining two or more apatite matrix flow-through fractions to form a purified biomolecule solution. In such methods, the biomolecule of interest in the purified biomolecule solution may comprise 90% or more of all biornolecule species in the solution, or may comprise 95% or more of all biomolecule species in the solution.

[0045] Collected apatite matrix flow-through fl-actions and purified hiornolecule solutions described herein may have a UV-visible spectrophotometric absorption ratio (280nm/260nrn) of 1.5 or more, preferably 1.7 or more.

[0046] Collected apatite matrix flow-through fractions and purified biomolecule solutions described herein may have a UY-visible spectrophotometric absorption ratio (2SOnmI26Onm) of 1.5 or more, preferably 1.7 or more; and in such fractions or solutions no contaminant proteins may he detectable on an SDS-PAGE gel run under reducing conditions and stained with coornassie blue dye, optionally wherein the SDS-PAGE gel is overloaded with sample.

[0047] In collected apatite matrix flow-through fractions and purified biomolecule solutions described herein, the biomolecule of interest may be present at a concentration of 0.4-0.7mg/mi.

[0048] In any of the methods defined herein the apatite matrix may he in the form of a column having an internal diameter to height ratio of 1 or more.

[0049] In any of the methods defined herein the apatite matrix may be in the form of a column having an internal diameter of approx. Scm and a height of approx. 7cm.

[0050] In any of the methods defined herein the apatite matrix is a hydroxyapatite matrix, optionally formed of either a derivatized or a non-derivatized hydroxyapatite, further optionally formed of a ceramic hydroxyapatite, more preferably wherein the hydroxyapatite matrix comprises porous spheres of non-derivatized ceramic hydroxyapatite of diameter approx. 40tm having pores of diameter of approx. 600-800 K [0051] The invention also pi-ovides for the use of an apatite matrix in any of the separation methods defined herein, optionally wherein the apatite matrix is described as above, preferably wherein the hydroxyapatite comprises porous spheres of non-derivatized ceramic hydroxyapatite of diameter approx 4olim having pores of diameter of approx. 600-800 A [0052] The invention also provides a method for producing a purified hiomolecule solution from a contaminated hiomolecule solution by separating a hiomolecule of interest contained in the contaminated biomolecule solution from one or more contaminant species contained in the contaminated biomolceule solution, the method comprising: a) performing a first separation step using an ion exchange matrix, the first step comprising: i. equilibrating an ion exchange matrix with an ion exchange matrix equilibration solution; ii. loading the equilibrated ion exchange matrix with a volume of the contaminated hiomolecule solution; iii. contacting the loaded ion exchange matrix with a volume of an ion exchange matrix wash solution; iv. allowing the loaded contaminated biomolecule soludon and the ion exchange matrix wash solution to flow through and exit the ion exchange matrix to form an ion exchange matrix flow-through; v. collecting fractions of the ion exchange matrix flow-through which contain the hiomolecule of interest and combining the collected fractions to form an intermediate biomolccule solution; and h) performing a second separation step using an apatite matrix, the second step comprising: i. loading an apatite matrix with a volume of the intermediate biomolecule solution, wherein the apatite matrix has been equilibrated with an apatite matrix equilibration solution; ii. contacting the loaded apatite matrix with a volume of an apatite matrix wash solution; iii. allowing the intermediate biomolecule solution and the apatite matrix wash solution to flow through and exit the apatite mati-i x to fonii an apatite matrix flow-through; iv. collecting fractions of the apatite matrix flow-through which contain the hiornolecule of interest and which are free of or substantially free of the one or more contaminant species; v. combining the collected apatite matrix flow-through fractions 1 0 to form the purified hi 011101 ecu Ic sol uti oil; and wherein: 1) upon contact with and transit through the equilibrated ion exchange matrix and the equilibrated apatite matrix the bioniolecule of interest and the one or more contaminant species do not become adsorbed to the matrices; 2) no desorption step is performed during either the first or second separation steps; 3) during the second separation step the hiomolecule of interest is separated from the one or more contaminant species following differential rates of transit of the biomolecule of interest and the one or more contaminant species through the apatite matrix.

In such a method, the contaminated biomolecule solution may be produced by the steps comprising: a) obtaining a volume of a crude preparation containing the biomolecule of interest, optionally wherein the crude preparation is a cell-free culture supernatant; h) performing a protein precipitation step. optionally using ammoiiium sulphate, preferably wherein the volume of the crude preparation is adjusted to be 65% saturated with ammonium sulphate; c) collecting precipitated proteins, dissolving the precipitated proteins in a resuspension solution, removing any 11011-dissolved proteins and collecting the dissolvate; and d) performing a dialysis step upon the dissolvate using a dialysis solution to form the contaminated biomolecule solution.

[0053] In the immediately above-described methods the contaminated hiomolecule solution, the ion exchange matrix equilibration solution, the ion exchange matrix wash solution, the intermediate hiomolecule solution, the apatite matrix equilibration solution, the apatite matrix wash solution, the purified hioniolecule solution, the resuspension solution and the dialysis solution may all he comprised of a buffered solution having the same composition [0054] In the irnmediae1y above-described methods the volume of the intermediate biomolecule solution is approx. 120m1 to 150rn1.

[0055] In the immediately above-described methods each buffered solution may comprise potassium phosphate pH7.2 at a concentration of between 1O-75rnIvI, preferably 50mM potassium phosphate pU72. Each buffered solution may further comprise DYE at a conceitation of between 0,5 -5 mlvi. preferably 1mM DII. Each buffered solution may comprise 50mM potassium phosphate pH7.2 and 1mM DTT.

[0056] In the relevant methods defined herein incorporating the use of an ion exchange matrix, the ion exchange matrix may comprise beads of cross-linked agarose further comprising functional groups of CH2N(CH and counter ions of P043 [0057] The invention also provides a kit of parts for use in any of the methods defined herein, the kit comprising a receptacle containing a quantity of an apatite matrix, preferahly any hydroxyapatite matrix as defined herein ahove, and a receptacle containing a quantity of an ion exchange matrix, preferably wherein the ion exchange matrix comprises heads of cross-linked agarose further comprising functional groups of CI-12N4(CI-13)3 and counter ions of PO4.

[0058] In such a kit, the mauix-containing receptacles may be in the form of chromatogi-aphic colunrns suitable for use together in a biomolecule sepai-ation method.

BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 shows SEQ ID NO: 1 which is the amino acid sequence of the deoxyribonuelcasc cnzyme NucB derived from Bacillus lichenifoi-rnis strain EI-34-6 (the sequence is also available as Genbank accession number ADX31661), The signal peptide sequence is shown underlined.

[0060] Figure 2 shows SEQ ID NO: 2 which is the nucleic acid complete cDNA sequence of the deoxyribonuclease enzyme nueB derived from Bacillus lichenifonnis strain EI-34-6 (the sequence is also available as Genbank accession number lIQI 12343).

[0061] Figure 3 shows an SDS-PAGE gel run under reducing conditions and subsequently stained with coomassie blue dye. The gel shows increasing loadings of the contaminated biomoleculc solution comprising ammonium sulphate precipitated material which was recovered from the cell-free culture supernatant, re-dissolved in SOrntv1 potassium phosphate pH7.2, 1mM DII and dialysed against the same buffer.

[0062] Figure 4 shows an SDS-PAGE gel run under reducing conditions and subsequently stained with coomassie blue dye. The gel shows individual fractions of material which flowed through a Q Sepharose'TM matrix column upon loading with a volume of the contaminated hiomolecule solution and subsequent application of a wash buffer comprising 50mM potassium phosphate p117.2. 1mM DTT [0063] Figure 5 shows an SDS-PAGE gel run under reducing conditions and subsequently stained with coomassie blue dye. The gel shows individual fractions of material which flowed through a hydroxyapatite matrix colunrn upon loading with a volume of the intermediate solution and subsequent application of a wash buffer comprising 50mM potassium phosphate pH7.2, 1mM DTT.

[0064] Figure 6 shows an SDS-PAGE gel run under reducing conditions and subsequently stained with coomassie blue dye. The gel shows increasing loadings of pooled fraction numbers 40 to 54 of material which flowed through the hydroxyapatite matrix column.

[0065] Figure 7 shows an SDS-PAGE gd run undcr rcducing conditions and subsequently stained with coomassie bluc dyc. The gel shows increasing loadings of pooled fractions of material which flowed through the hydroxyapatite matrix colunm in a repeat experiment.

[00661 Figure 8 shows overlaid flaces of the absorption spectra for three independent purifications of NucB using the separation method described.

DETAILED DESCRIPTION

[0067] Described herein are methods for purifying a biornolecule of interest by separating the biomolecule of interest from one or more contaminant species using an apatite niaftix. Also descnbed are methods for producing a solution containing a purified hiomolecule starting from a solution containing the hiomolecule of interest and at least one or more contaminant species, the methods involving the use of an apatite matrix. Uses of an apatite matrix in methods of separating a hiomolecule of interest from one or more contaminant species are also described, as are kits for use in such methods. In all methods described herein an apatite matrix is used to purify the biomolecule of interest by separating the biomolecule of interest from the one or more contaminant species by differential transit rate separation.

[0068] The methods described herein provide a means for separating a biomolecule of interest, usually a polypeptide or protein, from one or more contaminant species so as to produce a biomolecule preparation having a consistently high yield and a consistently high degree of purity wherein the hiomolecule, because of its characteristic hiophysical/physicochemical properties, can he difficult to purify by conventional separation methods and matrices. Typically the hiomolecule of interest is a hiomolecule that is refractory or difficult to he purified by its separation from one or more contaminant species using only an ion exchange material as the sole separation matrix. The biomolecule of interest may be a biomolecule that is refractory or difficult to be purified by its separation from one or uiore contaminant species using only a size exclusion material as the sole separation uiatrix such as a gel filtration uiatrix. The biouioleeule of interest may be a biomolecule that is refractory or difficult to be purified by its sepa-ation from one or more contaminant species using a combination of both an ion exchange maft-ix and a size exclusion maix, such as a gel filntion maix, or wherein the biomolecule of interest simply cannot (for practical purposes) he separated from the one or more contaminant species by such matrices because of its chai-aeteristic biophysica]Jphysicochemical properties.

[0069] Thus the invention provides a method of separating a biomoleeule of interest from one or more contaminant species comprising contacting a buffer-equilibrated apatite matrix with a sample containing the hiomolecule of interest and the one or more contaminant species, washing the matrix with an apatite matrix wash solution and collecting fractions of the apatite matrix flow-through which contain the biouiolccule of interest and wherein the biomolcculc of interest is separated from the one or more coiitarninant species; wherein upon contact with and transit through the apatite matrix the hiomolecule and the one or more contaminant species do not become adsorbed to the matrix and wherein no desorption step is performed; and wherein the hioniolecule of interest is separated from the one or more contaminant species following differential rates of transit of the hiomolecule and the one or more contaminant species through the apatite matrix [0070] The term "biomolecule" or "biornolecule of interest" as used herein refers to any biological molecule which is to be purified by separation from one or more contaminant species. Such a biological molecule will typically be a polypeptide or protein, but it could be another species of biological molecule such as a carbohydrate, polysaccharide or oligosaceharide etc. Preferred hiomolecules to he purified by the techniques described herein are proteins.

[0071] For the avoidance of doubt, unless the context clearly dictates otherwise references herein to singular terms such as "hiomolecule", "hiomolecule of interest", "protein", "contaminant species" etc. are intended to relate to the plural, e.g. a population of biomolecules etc. Unless the context clearly dictates otherwise, as used herein the term "approximately" is taken to mean +1-10% of the stated amount.

[0072] Where the biomolecule of interest is a polypeptide or protein, it may be a recombinant polypeptide or protein. Recombinant polypeptides or proteins may be expressed from a nucleic acid which has been created by recombinant techniques. A recombinant protein or polypeptide may typically be expressed from a gene which is not native to the cell in which it is expressed. often referred to as an exogenous gene.

A recombinant protein or polypeptide may also he the expression product of a gene which is native to the cell in which it is expressed but which gene has been structured in the genome of the cell in an artificial context by recombinant techniques. The biomolccule of interest may alternatively be a non-recombinant or native protein.

[0073] Terms such as "contaminant species", "contaminating species" and "contaminating hiomolecule" as used herein refer to any and all undesirable molecular species which are required to he separated away from the hiomolecule of interest to he purified. These will also typically be another polypeptide or protein. However, they may also he any other biological molecule such as a peptide. carbohydrate, saecharide, oligosaeeharide, polysaecharide, lipid. oligonucleotide. polynueleotide, nucleic acid, etc. Alternatively, they may not he a biological molecule. For example they may comprise any substance which has become undesirably associated with the biomolecule of interest to be purified, such as a non-biological component of a growth medium or a synthetic entity. Specifically with regard to the purification of the biornolecule known as NueB, as described in more detail herein, two species which consistently contaminated NucB preparations were a straw coloured material and a light scattering material. Neither of these species was properly defined as such, hut nevertheless they remained closely associated with NucB and therefore were significant contaminating species A "contaminant species", "contaminating species" or "contaminating hiornolecule" can therefore he any entity which undesirably associates with and is to he separated from the hiomolecule of interest to he purifie± [0074] A "one or more contaminant species". as described in the methods defined herein, will, like the biornolecule of interest, flow through an apatite matrix without adsorbing thereto. Such contaminailt species can therefore be collected together with the biornolecule of interest in a bulk apatite matrix flow-through sample and/or bulk apatite matrix wash sample. and will thereby contaminate the biornolecule of interest h the bulk sample. Other more general "contaminating species" may adsorb to an apatite matrix when contacted therewith Such contaminating species may therefore be sftaightforward to sepai-ate from the biornolecule of interest using apatite. For example, a biornolecule of interest (like the protein NucB as described herein) will not adsorb to a given separation matrix, such as an apatite matrix, whilst certain contaminating species may adsorb to that matrix. The differential binding to the matrix will then facilitate the separation of the specific contaminating species from the biomolecule of interest. Where a biomolecule of interest and certain contaminating species have one or more biophysicallphysicochernical properties which differ from each other, e.g. isoclectric point, their separation using a separation mathx may be facilitated However if a hioniolecule of interest and one or rnoi-e specific contaniiiiant species has one or more hiophysical/physicochemical properties which are similar, then it may be more difficult to sep' -ate them using a given separation maUix, e.g. if they show little or no significaiit differential hindiiig to the separation matrix mid co-elute iii a hulk flow-through sample subsequent to being contacted with the matrix.

The methods described hereiii surprisiiigly allow both fonfis of undesirable species (apatite adsorbing and non-adsorbing forms) to be separated from the biomolecule of interest by an apatite matrix due to the utility of differential transit rate sepai-ation principles described herein.

[0075] The technique described herein can in particular he applied to separate a hiomolecule of interest from one or more contaminant species wherein the hiomolecule of interest could not he separated from the one or more contaminant species using e.g. (1) a single step separation method involving only a size exclusion matrix such as a gel filtration matrix, (2) a single step separation method involving only an ion exchange matrix or (3) a two-step separation method involving a combination of a size exclusion matrix/gel filtration matrix and an ion exchange matrix.

[0076] Because of the complexity of the physical interactions between bioinolecules and separation matrices, is impractical to attempt to define the biomolecules of interest and undesirable species that might not be separable using a size exclusion matrix such as a gel filtration matrix and/or an ion exchange matrix (and thus define certain biomolecules of interest that may be amenable to purification using the methods described herein). However, whether a given biomolecule is unable to be separated from one or more contaminant species, as described in the methods defined herein.

using a size exclusion matrix such as a gel filtration matrix and/or an ion exchange matrix under a given set of conditions can he determined empirically by contacting a sample containing the biornolecule with such mathces and observing the binding behaviour of the biomolecule relative to the one or more contaminant species.

Biomolecules which are refractory or difficult to be purified by their separation from one or more contaminant species using one or both such matrices may he amenable to separation using the apatite-hased techniques described hei-ein.

[0077] The term "ion exchange niatrix" is a term of art and will be understood by one of skill in the art. Such a matrix will typically comprise a mass of material, often in the form of a resin or a gel-like substance and typically being of head-like structure.

Often materials such as cellulose or agarose are used to form the base resin/gel-like structure and further comprise charged functional groups which are covalently bonded to the base sneture. The functional coups provide the basis for charge-based interactions with other molecules.

[0078] The terms "size exclusion matrix" and "gel filtration matrix" are also terms of art and will be understood by one of skill in the art Such a matrix will typically comprise a mass of material, often in the form of small heads constructed of a polymer material and containing pores of differing sizes. The pores form the basis for differential separation of molecules hased on their size and/or molecular weight.

[0079] Similarly, an "aparite matrix" will be undcrstood to bc a mass of apatite rnatcrial adaptcd for the scparation of molecules of intercst. By apatitc it is meant that any suitable apatite mineral material can be used in the methods described herein, such as fluoroapatite and hydroxyapatite and composites thereof, Preferred methods and uses described herein involve hydroxyapatite.

[0080] As described earlier, because of its characteristic mixed mode mechanisms of binding to molecules, apatite materials, and in particular hydroxyapatite, have become regarded as specialised and non-routine separation materials. Such materials would not normally he investigated for the separation of a complex mixture. They are typically used to separate less complex mixtures, by allowing hiomolecules to become adsorbed to an apatite matrix, washing the matrix to remove unadsorhed species and then recovering the biomoleeule of interest in a more purified state by desorbing and eluting the biomoleeule of interest horn the matrix by use of suitable desorption buffers.

[0081] In contrast it has surprisingly been found that an apatite matrix, in particular a hydroxyapatite matrix, can sometimes be used to separate mixtures without any requirement for the hiomoleeule of interest and the one or more contaminant species to remain adsorbed to the matrix. Thus, described herein is a process by which an unadsorhed biomolecule of interest is separated from the one or more unadsorhed contaminant species by differential rates of transit through an apatite matrix, as described in more detail herein.

[0082] Various apatite materials are available for biornolecule separation purposes.

In particular, hydroxyapatite materials are available as native crystalline hydroxyapatite or in modified forms such as ceramic hydroxyapatite. For example various types of ceramic hydroxyapatite material are available from Bio-RadTM Laboratories, Inc. These have elution characteristics similar to crystalline hydroxyapatite, hut with various differences. Type I ceramic hydroxyapatite media comprises crystalline hydroxyapatite material which has been sintered at 400°C. and as a result has a higher protein binding capacity and a better capacity for the adsorption of acidic proteins. Type II ceramic hydroxyapatite media comprises material which has been sintcred at 700°C, and as a result has a lower protein binding capacity. However, it offers better resolution of nucleic acids and has been shown to be particularly useful for the purification of many species and classes of immunoglobulins. Both media are available in a range of pore sizes (currently 20, 40, and 80 Mm). Composite materials comprising hydroxyapatite and fluoroapatite are also available Macroprep Ceramic Hydroxyapatite, Type 1, 40Mm (available from Bio-RadTh) is a prefeiTed material.

[0083] The surfaces of apatite materials may he further optimised for the purification of biornolecules of interest by derivatization using species such as metal cations. With respect to the methods described herein, derivatization may improve the adsorption of certain contaminating species, whilst allowing the protein of interest to transit through the apatite matrix so as to separate the protein of interest from unadsorbed contaminating species. Thus any of the methods described herein may be performed using derivatized or non-derivatizcd apatite including dcrivadzcd or non-dcrivatizcd hydroxyapatite. The hydroxyapatite matrix may comprise porous spheres of non-derivatized ceramic hydroxyapatitc of diameter approx. 401tm having pores of diameter of approx. 600-800 A. [0084] Biomolecule separation by differential transit rate principles may be achieved using any of the above-described apatite materials, because of their broadly common characteristic mechanisms of hiomolecule interaction. However, it will he appreciated that unburdensome routine empirical testing may be required to arrive at the opdrnal material for the purification of a given biornolecule of interest, As noted, hydroxyapatite materials are the most prefened materials for performance of the methods described herein.

[0085] The separation matrices as used herein are in practice typically structured in the form of chromatographic columns These are typically hollow columns made of glass or other suitably inert material filled with a tightly packed hulk amount of the relevant separation matrix material and having entry and exit points at either end. One of skill in the art will readily understand how to produce such a structure in a manner suitable for the separation of biornolecules such as proteins. The specific dimensions of each colunm is not critical and can be adjusted depending upon the relevant circumstances, such as the amount of starting material involved in any given purification procedure. Consequently, the volumes of buffer required to equilibrate the separation matrix colunrn as well as the volumes of biomolecule-containing solution and wash solutions can be adjusted empirically depending upon the specific dimensions of the column. The volumes of fractions of flow-through solution (eluate) exiting the colunm can also be adjusted depending upon the specific dimensions of the column and volumes of load and wash buffers involved.

[0086] The terms "adsorb" and "adsorption" are terms of art and will he understood by one of skill in the art. They refer to the well-understood phenomenon by which enhties such as molecules, ions and atoms adhere to the surfaces of subshates and become fixed and inmiobilised thereto. In the contexts used herein the term typically desci-ihes the behaviour of hiornolecules or other entities in adhering to the surfaces of a separation matrix such as an ion exchange matrix or an apatite matrix, whereiii the attachment of the entity is sufficiently strong such that matrix wash buffers are unable to detach the entity from the matrix.

[0087] The terms "desorh" or "desorptioii" are also terms of art and will he understood by one of skill in the art. They refer to the well-understood phenomenon by which entities such as molecules, ions and atoms previously adsorbed to the surfaces of substrates become detached therefrom. In the contexts used herein the term typically describes the behaviour of biornolecules or other entities in detaching from the surfaces of a separation rnaix such as an ion exchange matrix or an apatite matrix.

[0088] The methods described herein involve the purification of a hiornolecule of interest from one or more contaminant species by differential transit rate separation using apatite and particularly hydroxyapatite. Differential transit rate separation involves a hiomolecule of interest and at least one contaminant species, both of which are unadsorbed to an apatite matrix under the conditions used. Upon application of a wash butler to an apatite matrix loaded with a biomolecule of interest and the one or more contaminant species, as defined herein, the biornolecule of interest and the one or more contaminant species will, under the action of gravity and/or another motive force, transit through the apatite manix. Differential uansit rate separation describes the phenomenon whereby the transit or migi-ation rates of the biomolecule of interest and the one or more contaminant species differ such that each species exits (or elutes from) the apatite manix at different times and thus can be collected in different fractions of the flow-through. Thus the one or more contaminant species may migrate through the apatite matrix at a faster rate than the hiomolecule of interest and may therefore exit the apatite matrix before the hiomolecule of interest or vice versa Thus the differential transit rates through the apatite matrix provide a means for separating the biornolecule of interest from the one or more contaminant species.

[0089] It has been further found that in certain situations good yield of the hiomolecule of interest, as well as good purity, in the apatite matrix eluates and pooled samples thereof can he achieved by a first separation step prior to the apatite matrix separation step. The first scpai-ation also involves a separation matrix, but one that is not apatite based. More specifically. it has been observed that the use of an ion exchange matrix in a prior separation step before the apatite matrix separation step can improve yield and purity by acting to remove certain other contaminating species before the apatite matrix separation step. thus reducing the complexity of the sample mixture which is subsequently applied to the apatite matrix. It will be appreciated.

however, that separation matrices other than an ion exchange matrix could be used to reduce the complexity of the sample mixture applied to the apatite matrix, depending on the biophysicallphysicochemical properties of the biomolecule of interest, For example, size exclusion matrices such as gel filtration matrices could he used. In the methods described herein the hiomolecule of interest will not become adsorbed to the separation matrix of the first separation step hut certain other contaminating species may become adsorbed thereto. Thus, certain other contaminating species may he retained in the separation matrix of the first separation step due to differences in size, molecular weight etc. As such, the separation matrix of the first separation step is used to fractionate the sample. Eluates from the separation matrix of the first separation step will be collected which contain the biomoiccule of interest. These eluatcs. however. may also contain the one or more contaminant species which also will not become adsorbed to the first separation matrix and will co-elute with the biomolecule of interest from the first separation matrix. Nevertheless, the first separation matrix will retain a proportion of other contaminating species present in the initial sample such that the eluates from the first separation matrix contain an increased relative proportion of the bioinolecule of interest. It will be appreciated, however, that in certain situations samples containing a biomolecule of interest which is to be purified may comprise a less complex mixture and may already contain a reduced relative proportion of undesirable contaminant and contaminating species, in which case the first separation step may he omitted and the use of apatite may comprise the only separation step involving a separation matrix.

[0090] Thus the invention also relates to a method for producing a purified hiornolecule solution from a contaminated hioniolecule solution by separating a hiornolecule of interest contained in the contaminated biomolecule solution from one or more contaminant species contained in thc contaminatcd biomolccule solution, the method compnsing: a) performing a first separation step using an ion exchange matrix, the first step cornpnsing: i. equilibrating an ion exchange matrix with an ion exchange matrix equilibration solution; ii. loading the equilibrated ion exchange matrix with a volume of the contaminated biomolecule solution; iii, contacting the loaded ion exchange maix with a volume of an ion exchange matrix wash solution; iv, allowing the loaded contaminated hiomolecule solution and the ion exchange matrix wash solution to flow through and exit the ion exchange matrix to form an ion exchange matrix flow-through; v, collecting fractions of the ion exchange matrix flow-through which contain the biornolecule of interest and combining the collected &actions to form an intermediate bioniolecule solution; and b) performing a second separation step using an apatite matrix, the second step comprising: i. loading an apatite matrix with a volume of the intermediate hiomolecule solution, wherein the apatite matrix has been equilibrated with an apatite matrix equilibration solution; ii. contacting the loaded apatite matrix with a volume of an apatite matrix wash solution; iii. allowing the intermediate hiomolecule solution and the apatite matrix wash solution to flow through and exit the apatite matrix to form an apatite mathx flow-through; iv. collecting fractions of the apatite matrix how-through which contain the biornolecule of interest and which are free of or substantially fiee of the one or more contaminant species; v. combining the collected apatite matnx flow-through fl-actions to form the purified hiomolecule solution; and wherein: 1) upon contact with and transit through the equilibrated ion exchange matrix and the equilibrated apatite matrix the biomnolecule of interest and the one or more contaminant species do not become adsorbed to the matrices; 2) no desorption step is performed during either the first or second separation steps; 3) during the second separation step the bioinolecule of interest is separated from the one or more contaminant species following differential rates of transit of the hiomolecule of interest and the one or more contaminant species through the apatite matrix.

[0091] The term "purified hiomolecule solution" as used herein refers to a buffered solution containing the hiomolecule of interest which has been separated from the one or more contaminant species, as described in thc methods defined herein. A purified biomolecule solution as used herein has been subjected to at least one separation step using a separation matrix which is an apatite matrix. In certain cases the purified bioinolecule solution as used herein has been subjected to a fhst separation step using a non-apatite separation matrix as described herein, preferably an ion exchange matrix, followed by a second separation step using an aparite matrix.

[0092] Other than the consdtuents of the buffer, the purified biornolecule solution will preferably contain only the bioinolecule of interest and no other species of hiomolecule. However, as would he immediately apparent to one of skill in the art since it may he difficult to entirely eliminate all molecules of the one or more contaminant species using the methods described herein, a minimum quantity of the one or more contaminant species may remain present and the solution may nevertheless still comprise a "purified biomolecule solution" according to these methods. Nevertheless, a "purified bioniolecule solution" as described herein will have been subjected to at least one separation step using a separation matrix which is an apatite matrix.

[0093] The term "contaminated biomolecule solution" or "sample" as used herein refers to a buffered solution containing the hiomolecule of interest, wherein the hiomolecule of interest has not been separated fl'om the one or more contaminant species, as defined. A contaminated biomolecule solution or sample as described herein has not been subjected to any separation step using a separation matrix.

Typically, contaminated biomolecule solutions or samples containing the biomolecule of interest and the one or more contaminant species will be complex mixtures containing a high proportion of many different undesirable species relative to the proportion of the hiomolecule of interest, such solutions or samples are typically formed from crude samples such as cell-free culture supernatants often further involving protein precipitation, resuspension and dialysis steps, as described in more detail below. Thus, a purified hiomolecule solution (and fractions of apatite matrix flow-through which contain the hiomolecule of interest and wherein the hiomolecule of interest is separated from the one or more contaminant species) will contain a highcr proportion of the bioniolceule of interest relative to the onc or more contaminant species than the proportion in the contaminated biomolecule solution.

[0094] The term "intermediate biomolecule solution" as used herein refers to a buffered solution containing the biomolecule of interest, wherein the biomolecule of interest has been partially separated. An intermediate biomolecule solution as used herein has been subjected to a first separation step using a non-apatite separation matrix, such as an ion exchange separation matrix, but has not been subjected to a second separation step using an apatite matrix and therefore retains one or more contaminant species.

[0095] In methods involving a first separation step using a non-apatite separation niatrix a preferred separation matrix is an ion exchange matrix. A particularly preferred ion exchange matrix comprises beads of cross-linked agarose further coniprising functional groups of CH2N(CH)3. Such a matrix is commercially available under the trade name Q Sepharosetm. It will he appi-eciated that other forms of ion exchange matrix may he used depending upon the paiticular hiornolecule of thtcrcst to bc punficd and routinc optimisation can be employed to determine the optimal form. In methods used herein such a matrix may comprise counter ions of PD43-.

[0096 At various steps of the methods described herein buffered solutions are used.

Buffered solutions described herein are an ion exchange matrix equilibration solution, an ion exchange matrix wash solution, an apatite matrix equilibradon solution and an apatite matrix wash solution. Also described is a resuspension solution and a dialysis solution. The contaminated biornolecule solution, the sample, the intermediate hiomolecule solution and the purified hiomolecule solution also are all comprised of a buffered solution. The buffered solutions may he constituted with specific reagents and adjusted to a specific p1-I depending upon the particular hiomolecule of interest to he purified. The buffered solutions may comprise potassium phosphate p117.2 at a concentration of between 10-75mM, preferably the solution may comprise 50mM potassium phosphate p117.2; optionally each buffered solution may further comprise DTT at a concentration of between 0.5-5.0mM, preferably iniM DTT. A preferred buffered solution comprises 50mM potassium phosphate pH7.2, 1mM DTT. One significant advantage of the procedures described herein is that the buffered solutions used throughout the procedure may be constituted of the same reagents and pH and therefore no changes in buffer composition are required. This significantly simplifies the separation procedures and saves time and cost, A preferred buffered solution used in all steps of the methods described herein comprises 50mM potassium phosphate pH7.2, 1mM DTT. However, where it is desired to vary the constitution of one or more buffers, one of skill in the art will be able to do so in a routine and empirical way so as to achieve conditions whereby differential transit rate separation using an apatite matrix as described herein maybe performed.

[0097] The techniques used herein can provide relevant samples of apadte matrix flow-through fractions (eluates) and a purified biomolecule solution which contain the biomolecule of interest to be purified, and wherein the biomolecule of interest is separated from the one or more contaminant species, as described iii the methods defined herein. Preferably all of the one or more contaminant species which are prcsent in the samplc applied to or loaded onto the apatite matrix arc abscnt from thc purified hiomolecule solution and from the relevant apatite matrix flow-through fractions or pooled combinations thereof. In these samples, apart from the hiomoleeule of interest, preferably no contaminant proteins are detectable on an SDS-PAGE gel run under reducing conditions and stained with coomassie blue dye. In these samples, apart from the bioinolecule of interest, preferably no contaminant proteins are detectable on a polyacrylarnide gel overloaded with sample and stained with eoomassie blue dye. In the relevant samples of apatite matrix flow-through fractions (eluates) and in the purified biornolecule solution which contain the hiomolecule of interest to he purified, as described above, the liv absorption ratio (280nm/260nm) may he 1-5 or more, preferably 1.7 or more (indicating lack of contamination with nucleic acids).

[0098] In samples of apatite matrix flow-through fractions (eluates), or in pooled combinations thereof, or in a purified hiomolecule solution which contain the biomoleeule of interest to be purified and wherein thc biomolecule of intercst is separated from the one or more contaminant species, the biomolecule species of interest preferably comprises 90% or more of all biomolecule species in the solution.

The biornolecule species of interest preferably comprises at least 90% but preferably at least 95%. In these samples the biomoleeule species of interest is present at a concentration of 0.4-0.7mg/mi. preferably 0.5mg/mi.

[0099] Starting from 1OL of cell-free cultme supernatant it has been found that the technique can routinely provide a bulk purified biomolecule solution which comprises between 50-80mg of protein.

[00100] Particularly preferred hiomolecules of interest which can he purified by apatite-based differential h'ansit rate separation are the class of secreted nuclease enzymes, particularly secreted nuelease enzymes of microbial origin. More particularly, the deoxyribonuelease enzyme NueB derived from Bacillus lichen jtbrrnis, and i-elated polypeptides, al-c amenable to purification by the methods described herein. NueB as exemplified in the examples herein is derived from Bacillus licheniforunis strain EI-34-6, the amino acid sequence of which is listed as SEQ ID NO: I as shown in Figure 2 (and which corresponds to Genhank accession number l-IQ1 12343).

[00101] Any enzymatic ally-active truncate, fragment, homolog, ortholog, paralog.

analog or derivative of a microbial NucB deoxyribonuclease enzyme may be purified by the apatite-based differential transit rate sepai-ation techniques described herein, including mutated or genetically altered forms.

[00102] It has been found that a desirable method for achieving a final preparation of S purified NucB having a high yield and purity and separated from all or substantially all of the one or more contaminant species involves a technique whereby an initial contaminated hiomolecule solution is subjected to a first separation step using an ion exchange matrix as detailed herein, followed by a second separation step using an apatite matrix as detailed herein Preferably the ion exchange matrix comprises beads of cross-linked agarose further comprising functional groups of CH2N4(CH)3 and counter ions of P043. Preferably each bufidred solution in the method comprises 50mM potassium phosphate pH7.2, 1mM DTT.

[00103] With regard to the production of a final preparation of purified NucB as described above, a preferred method of producing an inidal contaminated biomolecule solution which is to he applied to the ion exchange matrix is descrihed The preferred method involves obtaining a volume of a crude preparation containing NucB; performing a protein precipitation step. preferably using arnmnonium sulphate, preferably wherein the volume of the crude preparation is adjusted to he approx at least 65% saturated with ammonium sulphate; collecting precipitated proteins; dissolving the precipitated proteins in a resuspension solution, removing any non-dissolved proteins and collecting the dissolvatc; and performing a dialysis step upon the dissolvate using a dialysis solution. In these methods the resuspension solution and the dialysis solution are preferably comprised of a buttered solution having the same composition as the buffered solutions used in the subsequent matrix-based separation steps, preferably the i-esuspensioii solution and the dialysis solution comprise 50mM potassium phosphate pH7.2, 1mM DTT.

[00104] In the method described above the ci-ude preparation contain rig NucB may preferably he a cell-fl-ce culture supernatant formed as a result of growing microorganisms which express the NucB enzyme and wherein the NueB enzyme is secreted into the culture growth medium. An example of such a method has been described previously 151. The cell-free culture supernatant is formed by removing the microorganisms from the growth medium, normally by centrifugation. Whilst expression of NucB fi-om microorganisms is a preferred method, it will be appreciated that a crude preparation containing NucB may he obtained by a variety of means, such as by en vitro transcription/translation reactions 00105] For the purification of the deoxyrihonuclease enzyme NucB derived from Bacillus lieheniforinis, a preferred apatite material for the second separation step comprises porous spheres of ceramic hydroxyapatite (non-derivatized), wherein the spheres have a diameter of approx. 4Ojrni and the pores have a diameter of approx.

600-800 A (Macroprep Ceramic Hydroxyapatite, Type 1, 4Opm available from Bio-Rad). although again it will be appreciated that other apatite materials may be of use and can be determined routinely and empirically.

[00106] The separation techniques described above have been established on the basis of a desire to purify, essentially to homogeneity, the secreted exil-acellular deoxyribonuclease enzyme NucB from Bacillis 1ichenfo,-inis, The technique was developed because lab-scale attempts using a variety of conventional separation methods had pi-oved i-elatively unsatisfactory iii sepai-atiiig NueB in the desired way [00107] A one-step size exclusion (gel filtration) chromatography procedure per se had been unsatisfactory in providing the degree of separation required [5]. Without wishing to be bound by any particular theory. this suggests that certain contaminant species present in the initial contaminated biornolecule solution migrated through a colullm of a gel filtration matrix at the same transit rate as NucB. This would suggest that such contaminant species were of similar size or possessed similar biophysicallphysicocheniical properties as NucB. such that their migration behaviour through the colullm could not be distinguished from that of NucB. Alternatively, the transit rate of a population of NucB protein molecules through a gel filtration colunrn may not have been unifonn, and this may have inhibited separation to the required degree.

[00108] In addition, it was found that the NucB protein has a neuu-al charge at physiological pH (pH7). Again without wishing to be bound by any particular theory, this may account, at least in part, for the inability of NucB to adsorb to an ion exchange matrix. Thus the sole use of an ion exchange matrix per se had also proved unsatisfactory in providing the required degree of separation of NucB from other relevant contaminant species either by adsorption/desorption separation methods or by separation methods involving fractionation. This suggests that certain contaminant species present in the initial contaminated hiomolecule solution also possessed a neutral or net positive charge at physiological p1-I, and may have been unable to adsorb to the ion exchange matrix for that reason. Such contaminant species may furthermore also possess biophysical/physicochemical properties which would render their rniation behaviour through the ion exchange column similar to that of NucB, such that sepai-ation to the required degree could not be achieved, Thus, because of its neuUal charge. it was not possible to apply conventional ion exchange methods for the separation of NueB involving adsorption to the maix. extensive washing and then desorption from the matrix by an ion exchange step, and because of the behaviour of relevant one or more contaminant species relative to NucB, the use of an ion exchange matrix in a fractionation-based separation proved difficult.

[00109] It is believed that the techniques described herein could he applicable to the purification to homogeneity of other hiornolecules. particularly proteins, in a manner analogous to the purification of NucB as described herein. The technique may find utility for the purification of a protein which cannot be separated fioni one or more contaminant species by the sole use of a size exclusion matrix (e.g. gel filtration mati-ix), the sole use of an ion exchange matrix or a combination of both types of rnati-ices Whether any given protein would he amenable to purification in this way would be dependent upon several factors. For instance, the molecular weight and charge/isoelectric point of the protein to he purified i-dative to the molecular weight and charge/isoelectric point of the one or more contaminant species may dictate whether the described technique could he used, since these pi-operties will likely intluence whether the one or more contaminant species would consistently co-elute with the biornolecule of interest from a gel filtration and/or ion exchange column.

[001101 In any event, whether any given bioniolecule of interest would be amenable to purification in the described way could be established in an empirical, routine and unhurdensome manner by observing the hehaviour of the hiomolecule of interest when S interacting with separation matrices, and thus could he established without prior knowledge of the specific hiophysical/physicochemical properties of the hiomolecule of interest to he purified and the one or more contaminant species relative to the hiornolecule of interest [00111]The technique may for instance be applicable to a protein which, like NueB.

would possess a neutral charge at a pH of approximately 6.8-7.2 and which might be unable to adsorb to an ion exchange matrix, or which might not be able to he separated from one or more contaminant species by the sole use of an ion exchange matrix.

Such a protein may have the same or similat-moleculat-weight to NucB, approx.

12kDa.

[00112] The technique might be particularly applicable to a protein of this type which, like NucB, might not be separated from the one or more contaminant species by fractionation using an ion exchange matrix andlor by fractionation using a size exclusion matrix such as a gel filtration matrix.

[00113] In order for the described technique to be applicable for the purifieahon of a biomolecule, the biomoleeule must be capable of migrating through an apatite (preferably hydroxyapatite) matrix without adsorbing to the apatite mathx since this forms the basis of differential transit i-ate separation as described heiein. The amino acid sequence of the hiomolecule of interest should therefore pieferahly not have clustcrs of acidic amino acids containing more than two consecutivc aspartate or glutamate residues. This is because, as noted above, carhoxyl groups act in 2 ways with iespect to their inteiaction with hydroxyapatite -they are iepelled electrostatically by the negative charge of the crystalline material, and they bind specifically by complexing with the Ca2 sites of the material. The initial attraction between COO and Ca2 sites is electrostatic but the actual binding of the COO to the Ca2 sites involves the formation of a much stronger coordination complex. Hence retention on the column requires the cooperative interaction of several cai-boxyl coups. Clusters of carboxyl coups therefore facilitate most effective binding to hydroxyapatite and therefore should he avoided if the protein is to he amenable to he purified by differential transit rate separation However, it will he appreciated that the precise amino acid sequence of a protein may not always he determinative of the functional behaviour of a protein in an apatite matrix. The functional ability of a protein to migrate through an apatite (preferably hydroxyapatite) matrix without adsorbing to the apatite matrix can of course he determined empirically in a routine way.

EXAMPLES

Example 1

Rationale for the optimisation of protein purification methods OO114j The biomoiccule chosen for purification optiinisation was the extracellular deoxyribonucicase enzyme NucB of Bacillus lichen ifbrmis. Previously, a purified NucB protein preparation had been achieved which proved satisfactory to provide a source of deoxyrihonuclease enzymatic activity capable of biofilrn dispersal in a range of situations 151 In these previous studies, overexpression in Bacillus su.btiiis of the NucB enzyme was achieved using a suhtilin hduction system. Proteh was secreted into the culture medium, was collected as a supernatant and was concentrated following TCA precipitation. As described previously 5j, concentrated protein supernatants were then suhjected to bioassay-guided fractionatioit Multiple gel filtration (size exclusion) media were utilised (e.g. Sephadextm1 G-50, Sephadex'TM LH-20, Superosetm1 12) with the best separation in these studies being achieved using TCA precipitarioll followed by fractionation using Supcrose'TM 12 gel filtration [5].

However, whilst the preparation achieved following Supcrose'TM 12 gel filtration provided satisfactory deoxyribonuclease enzymatic activity, the preparation remained slightly contaminated with other biomolecule species. In addition, the previous technique did not provide consistency with respect to the amount of NucB protein recovered. It was consequently desired to provide an improved preparation containing NucB protein at a higher degree of purity and to identify a superior separation technique which could provide a greater degree of consistency in terms of the amount of protein recovered. Various further purification options were available. However, a technique which was found to he surprisingly superior in terms of final purity and consistency in recovered amounts of protein was determined as detailed herein.

[00115] In particular, the inventor developed a technique involving a specific combination of separation reagcnts and novel usc thereof, which had not been employed before to purify NucB, and, it is believed, which has not been employed before to purify any biomolecule in the manner described, Furthermore, in attempting to achieve a high yield and purity of NucB the inventor discovered that the separation reagents, and in particular the apatite reagent, could he employed in what is believed to be a novel method of use in protein separation science. The unpredictable combination of (a) the choice of specific separation reagents and (b) their particular method of use unexpectedly delivered NucB protein preparations of extremely high purity and achieved a high degree of consistency as between different purification batches.

[00116]It is understood that the techniques described herein can be employed to purify, in an analogous manner, proteins which have certain hiophysical/physicoehemical properties similar to NucB, and that therefore the technique described herein may have broad utility in protein separation methods.

Example 2

Production of a contaminated biomolecule solution Fool 17] In the studies detailed below, NucB protein was expressed in Bacillus subtilis using a subtilin induction system essentially as previously described [5]. The culture was subjected to centrifugation to form a cell-free culture supernatant.

FOOl 18] Approx. 1 OL of NucB-containing cell-free culture supeniatant was pooled and then split into 3 x SL beakers. The content of each beaker was adjusted to be 65% saturated with anirnonium sulphate whilst stirring in an ice/water mix. The fluffy white precipitate that forms on the top of each culture supernatant was removed and discat-ded and then the beakers were stoi-ed at 5°C ovet-night to allow quantitative precipitation of the NucB protein.

[00119] Following pi-ecipitation, the proteins weie subjected to centrifugation (SLA 3000 i-otor, 9k rpm. 4°C, 60 mm), collected and then resuspended in a minimum volume of SOmIVI potassium phosphate pH7.2, (50mM KP pH7.2) imIVI DII (10 ml for each of 12 centrifuge bottles of approximate capacity 500 nil). The mixture of proteins was stirred for 30 mm at 5°C. The stirred protein mix was then spun (Sorvall SS34 rotor, 14k rpm, 4°C. 30 ruin) to remove any non-dissolved proteins.

[00120] The supernatant was taken and dialysed against 3 x SL changes of 50mM KP pl-l7.2, 1mM DIT. Dialysis removes ammonium sulphate. peptides of less than 3kDa and other low molecular weight species. The resulting solution formed following dialysis was examined and was found to contain a large number of undesirable hiomolecules. The resulting solution formed at this stage is an example of what is referred to herein as a contaminated biomolecule solution.

[00121] Increasing loadings of the contaminated biomolecule solution produced following ammonium sulphate precipitation and dialysis is shown in Figure 3. Figure 3 shows an SDS-PAGE gel (run under reduchig conditions) treated with coornassie blue dye which stains proteins. The darkest band, which runs at the bottom of the gel, is NucB. However, as can clearly be seen, multiple undesirable protein species are present in the contaminated hiomolecule solution and which run above NucB on the gel.

Example 3

Production of an intermediate biomolecule solution [00122] As noted earlier, previous attempts using gel filtration matrices had not proved satisfactory in separating NucB to the desired level of homogeneity, and had proved relatively inconsistent in terms of the amount of protein recovered from batch to batch [5]. A separation method using an ion exchange matrix was considered.

[00123] The dialysed protein sample (contaminated hiomolecule solution) described in Example 2 was passed over a column comprising an ion exchange matrix (Q SepharosetmT). The colunrn (which in this case was structured to be 15 x 5 cm) had been equilibrated with 50mM KP pI-l72. 1mM DTT and was washed also with 50mM KPpH7.2, 1mM DYE [00124] Following analysis, it was found that the NueB protein did not adsorb to the ion exchange column. Analysis of the bulk ion exchange how-through revealed that NucB instead washes through the column. Consequently, it was not possible to separate NueB by conventional methods of ion exchange chromatography, that is to say by adsorbing NucB to the matrix, extensively washing the matrix and then specifically purifying NucB by collecting the protein following an ion exchange desorption step.

[00125] Despite the inability of NucB to adsorb to the column, the content of the Q SepharoseTh matrix eluate was thvestigated in more detail. The load and wash volumes were collected as 10 ml fractions after the column had been subjected to a 800nil volume of wash buffer comprising 50mM KP pH7.2, 1mM DII. NucB was typically found in fractions 27-60. Representative Q SepharoseTh fractions in this range are shown in Figure 4 which shows an SDS-PAGE gel (run under reducing conditions) treated with coomassie blue dye. The darkest hand, which runs at the bottom of the gel, is NucB. By comparison with Figure 3 it can he seen that although NucB does not itself adsorb to the ion exchange column, a significant proportion of the undesirable proteins present in the initial contaminated hiomolecule solution have nevertheless been removed from the Q SepharoseTh flow-through fractions by retention in the ion exchange matrix. Therefore some degree of separation can he achicvcd by this ion exchange matrix-assisted fractionation stcp.

[00126] However, although thc ion cxchange niatrix removcs a proportion of contaminating proteins from the contaminated biomolecule solution, protein contaminant species which could not be separated from NucB nevertheless remain in the Q SepharoseTM flow-through as can be seen from the coomassie-stained gel shown in Figure 4.

[00127] Furthermore, the NucB-containing Q Sepharosetm flow-through fractions were also significantly contaminated with a non-defined straw coloured material which also could not he separated from NucB. An additional component which also washes through the ion exchange matrix is a further non-defined substance that causes a high degrce of light scattering when the solution is analysed by UV and visible light spectroscopy. This light scattering material consequently also contaminates the NucB-containing Q Scpharose'M flow-through fractions.

[00128] Thus it was found that the Q SepharoseTM ion exchange matrix removes from the contaminatcd biomolcculc solution nucleic acids, certain other pcpddcs and components remaining in the growth medium at the end of the initial B. sn/ni/is fermentation step. as well as other non-NucB proteins secreted into the culture medium by the growing B. sithti/is cells. However, the iou exchange matrix was not capable of separating NucB in the contaminated biomolecule solution from at least the non-defined contaminating straw coloured material, the light scattering material and several other undesirable protein species. Furthermore, as noted above, since NucB failed to adsorb to the ion exchange matrix it was not possible specifically to purify NucB via a conventional method involving adsorption, colunm washing and a subsequent ion exchange desorption step.

[00129] It was therefore unclear how NucB could be purified to homogeneity as desired. Nevertheless, for further processing the appropriate NucB-containing fractions were pooled to form an example of what is referred to herein as an intermediate hiomolecule solution.

Examiile 4 Initial purification attempts using hyciroxyapatite [00130] As can be seen from Figure 4. a significant number of undesirable protein species remained in the ion exchange flow-through (intermediate hiomolecule solution). In addition, as described above, the intermediate biomolecule solution was also contaminated with a non-defined saw coloured material and a non-defined light scattering material. Like NucB, these species appeared unable to adsorb to the ion exchange matrix under the conditions tested and thus washed through the matrix to contaminate NucB in the intermediate hiomolecule solution.

[00131] It was clearly desirable to separate NucB froni the remaining contaminant species present in the intermediate biomoleculc solution. However, since the remaining contaminant species, like NucB, were unable to adsorb to the ion exchange matrix under the conditions tested, and since they co-eluted with NucB in the ion cxchangc flow-through, it was not clear how purification of NucB to homogcncity in the desired way could be achieved. It was clear that the intermediate hiomolecule solution remained a complex mixture of hiomolecules and the final separation of NucB from these remaining hioniolecule contaminant species was not straightforward.

Unconventionally, the use of a hydroxyapatite matrix to address this complex intermediate biomolecule solution was next investigated.

[00132] A colunrn of a hydroxyapatite matrix was established having in this case an internal diameter of approx. Scm and a height of approx. 7cm which gives an approx.

packed column volume of l37m1.

[00133] The hydroxyapatite column was equilibrated with 50mM KP p1-17.2. 1mM DTT. Following equilibration, approx. 120-1 SOmI of the intermediate hiomolecule solution comprising the pooled NucB eluate (also comprising 50mM KP p117.2, 1mM DTT) from the Q SepharoseTM ion exchange step was applied to the hydroxyapatite colijnm. Thc hydroxyapatite column was then washed with 700 nil 50mM KP pH7.2, 1mM DTT.

[00134] When a bulk hydroxyaparite flow-through was analysed it was found that, as with the Q SepharoseTh ion exchange matrix, NueB also did not adsorb to the hydroxyapatite matrix, but washed through the column. This precluded NucB from being purified also via a conventional method involving adsorption to the hydroxyapatite matrix colunrn, extensive colunrn washing and a subsequent desotption step. Crucially, both the sU-aw coloured material and light scattering material also did not adsorb to the hydroxyapatite matrix, and therefore contaminated the bulk flow-through sample containing NucB.

[00135] Thus attempts to purify NueB protein to homogeneity in the desired way using various conventional protein separation techniques has proved unsatisfactory. Various gel ifitrahon matrices had been attempted, as described previously [5], but none had proved capable of achieving the purity and yield consistency required. Since NucB failed to adsoih to an ion exchange matrix, it was not possible to perform adsorption, column wash and subsequent dcsorption steps and thus purify the protein by conventional ion exchange chromatography. Furthermore, since NucB failed to adsorh to a hydioxyapatite matrix it was also not possible to specifically purify the protein via a conventional mechanism involving desorption from a hydioxyapatite column either. Thus at this point it was wholly unclear how a workable chrornatogi-aphic methodology could be developed which might allow NucB to be separated from the various remaining contaminant species in the desired way.

Example S

Production of a purified biomolecule solution 00136] It is to he recalled that, like NucB, none of the protein contaminants remaining in the intermediate hiomolecule solution, the non-defined straw coloured material contaminant or the non-defined light scattering material conrnminant was capable of adsorbing to either an ion exchange matrix or a hydroxyapatite matrix under the conditions tested. Moreover, several of the contarninaiits co-eluted with NucB tioni both rnaices. rendering their sepai-ation difficult.

[00137] Unexpectedly, upon detailed and unconventional investigation of fractions of the hydroxyapatite matrix flow-through obtained in accordance with Example 4, it was found that NucB could indeed be separated from the remaining contaminants, as described below.

[00138]Thus. as before, a hydroxyapatite column was equilibrated with 50mM KP p1-17.2, 1mM DTL Following equilibration, approx. 120-150rn1 of the intermediate biomolceule solution comprising the pooled NucB eluate (also comprising 50mM KP pH7.2, 1mM DTT) from the ion exchange step was applied to the hydroxyapatite colullm. The hydroxyapatite column was then washed with 700 ml 50mM KP pH7.2, 1mM DTT. The load and wash volumes (hydi-oxyapatite matrix flow-thi-ough) were collected iii Grnl fractions for fui-ther analysis.

[00139] Surprisingly, despite the fact that neither NucB nor at least the non-defined straw colouied material and light scattering material coiitaniinants could adsorb to the hydroxyapatite matrix, it was found that these contaminants separated into different fractions of the hydroxyapatite matrix flow-through, and consequently NucB could be separated from these contaminant species.

[00140] The absoipdon spectrum (range 400 -240 nm) of the fractions from the hydroxyapatite elution profile were measured and analysed for the presence of NucB and the light scattering material. NucB was found typically in fractions 30-50.

S Typically, fraction numbers 1-26 were found to contain the straw coloured material contaminant. The light scattering material contaminant was typically found in fractions located at the end of the series of fractions containing the straw coloured material and at the beginning of the NucB-containing fractions.

[00141]Thus. completely uncxpcctcdly, it was found that NucB could be separated froni contaminant biomoleculc species, which species (like NucB itself) could neither adsorb to an ion exchange matrix nor a hydroxyapatite matrix and which could not be separated by fractionation using an ion exchange matrix. This revealed what is believed to be a hitherto unknown and unappreciated capability of a hydroxyapatite matrix in terms of its ability to allow the transit of a variety of unadsorbed hiomolecules of differing species through the matrix at different transit rates. As such.

the hydroxyapatite matrix can allow for the separation of a biomolecule species by transit rate fractionation, whereas separation could not be achieved by other more conventional modes such as size exclusion chromatography or by the use of an ion exchange matrix.

[00142] Spi volumes from consccutivc protein-containing frachons that were not straw coloured were run out on 15% SDS-PAGE under reducing conditions. The results are shown in Figure 5. As is clcar from a comparison between the gels of Figures 5 and 4.

the protein contaminant species present in the ion exchange fractions (Figure 4) are absent in later fractions from the hydroxyapatite flow-through (Figure 5).

[00143]The NucB-containing fractions were then pooled as appropriate, taking into account the data relating to the light scattering material, to form the final purified hiomolecule solution. Figures 6 and 7 show increasing loadings of relevant pooled hydroxyapatite flow-through samples (purified biomolecule solution) from independent purification runs. No visible protein contaminant species are observed.

[00144] Figure 8 shows overlaid traces of the absorbance specUit for thi-ee purifications of NucB using the separation method described. Each trace is from a separate and independent purification run and is the absorhance spectrum of a sample of a purified hiomolecule preparation of NueB produced following the pooling of relevant fractions of a hydroxyapatite flow-through as described above. It can he seen that (a) the A280/A260 ratio varies between 1.73-1.78 and (h) in the region 400- 3lOnm the spectra are super-imposable. In combination with analysis by SDS PAGE, this demonstrates the high degree of consistency between purification runs in terms of sample purity.

[00145] Analysis by differential scanning calorimetry of NucB purified as described above from separate and independent purifications showed that the biophysical properties of the finally purified NucB in the purified biomolecule solution had minimal variation from batch to batch.

[00146] Using the technique described, NucB protein was consequently collected free of contaminant species to a level of >95% purity. The NucB pool formed in this way from the starting volumes described is typically around 0.5 mg ml with a yield of approx.50-80 mg total starting from I OL of Bacillus subtilis cell-free culture supe rn atant.

[00147] In summary, the technique detailed herein has a number of significant advantages. The technique provides for the production of a purified bioniolecule solution (in this case of the protein NucB) wherein: * the protein could he consistently separated to a level of >95% purity; * the yield of the recovered protein was consistently high; and * the hiophysical properties of the purified protein had minimal variation from batch to batch.

In addition, and very importantly, because no adsorption/desorption steps are performed throughout the enthety of the method, the technique required no changes in buffer composition. Thus from the point of resuspension of the ammonium sulphate-precipitated protein (forming the contaminated hiomolecule solution) to the final collection of pooled fractions of the hydroxyapatite eluate (forming the purified hiomolecule solution) it was possible to use a buffer of the same composition throughout. This provides a method with reduced complexity and consequent savings in terms of time and cost.

[00148] It will be apparent that thc key observation which allowcd thc successful purification of NucB to the observed high level of purity, and with such consistency, was the realisation that the apatite matrix could be used in an unconventional method of differential nansit rate separadon, It will be appreciated that such a method could be used to purify other biomolecules and proteins whose behaviour in separation matrices is similar to that of NucB relative to one or more contaminant species.

Optionally, a prior initial separation step using another non-apatite matrix may additionally be desirable, such as the use of an ion exchange matrix as described herein.

REFERENCES

1. Gorbunoff, M. J. The Interaction of Proteins with Hydroxyapatite -I. Role of Protein Charge and Sflucture. Analytical Biochemistry; 136, pp425-432 (1984).

2. Gorbunoff, M. J. The Interaction of Proteins with Hydroxyapatite -II. Role of Acidic and Basic Groups. Analytical Biochemistry; 136, pp433-439 (1984), 3 Gorhunoff, Mi. and Timasheff, S N. The Interaction of Proteins with 1-lydroxyapatite -III. Mechanism. Anab'tical Biochenistr'y; 136, pp44O-445 (1984).

4. Kawasaki, T. Hydroxyapatite as a liquid chroniatographie packing. Journal of Chromatography A; 544(17), pp 147-1 84 (1991).

5. Nijland, R. et al. Dispersal of Bioffirns by Secreted, Matrix Degrading, Bacterial DNase. PLoS ONE; 5(12), ppl-7

Claims

CLAIMS1. A method of separating a biomolecule of interest from one or more contaminant species comprising contacting a buffer-equilibrated apatite matrix with a sample containing the biomolecule of interest and the one or more contaminant species. washing the matrix with an apatite matrix wash solution and collecting fractions of the apatite matrix flow-through which contain the biomolecule of interest and wherein the biomolecule of thterest is separated from the one or more contarnimmt species; wherein upon contact 1 0 with and transit through the apatite matrix the biomolecule md the one or more contaminant species do not become adsorbed to the matrix and wherein desorption step is performed; and wherein the hiomolecule of interest is separated from the oiie or more contaminant species following differential rates of ti-ansit of the hioniolecule and the one or more contaminaiit species 1 5 through the apatite matrix.

1 The method of claim 1 whereiii the hiomolecule of interest is a hiornolecule that cannot be separated from the one or more contaminant species using only an ion exchange matrix.

3. The method of claim I or claim 2 wherein the hiornolecule of interest is a biomolecule that cannot he separated from the one or more contaminant species using only a size exclusion matrix including a gel filtration maftix.

& The method of any of claims I to 3, wherein the apatite matrix equilihration solution, the sample containing the hiomolecule of interest and the one or more contaminant species, and the apatite matrix wash solution are all comprised of a buffered solution having the same composition.

5. The method of any of claims 1 to 4 wherein each buffered solution comprises potassium phosphate pH7.2 at a concentration of between 10-75mM, preferably 50mM potassium phosphate p117.2; optionally wherein each buffered soludon further comprises DTT at a concentration of between 0.5- 5mM, preferably 1mM DTT.

6-The method of any of claims 1 to 5, wherein the hiomolecule of interest is a polypeptide, optionally a recombinant polypeptide.

7. The method of claim 6 wherein the polypeptide has a neutral charge at p1-I 6.8-7.2, optionally wherein the polypeptide has a molecular weight of approx.l2kDa.

8. The method of claim 6 or claim 7 wherein the polypeptide is a microbial deoxyribonuclease enzyme, preferably a secreted microbial deoxyribonuclease enzyme.

9 The method of claim 8 wherein the microbial deoxyribonuclease enzyme is selected from the list of bacterial deoxyribonuclease, grain positive bacterial deoxyribonuclease. class Bacillus bacterial deoxyribonuclease, Bacillus lu7hentfonnts bacterial deoxyrihonuclease, Bacillus lichenifonnis strain FI-34- 6 bacterial deoxyrihonuclease, microbial NueB deoxyrihonuclease, or any enzymatically-active truncate, fragment, homolog, ortholog, paralog, analog or derivative thereof.

10. The method of claim 9 wherein the deoxyribonuclease polypeptide is Bacillus lieheniforinis strain El-34-6 NucB deoxyrihonuclease defined by SEQ ID NO: 1. or any enzymatically-active truncate, fragment, homolog, ortholog. paralog, analog or derivative thereof 11. The method of any of claims I to 10 wherein the one or more contaminant species is a hiomolecule or hiomoleeules which are selected fiom the list of peptides. polypeptides. proteins, carbohydrates, saccharides, oligosaeeharides.polysaccharides, lipids, oligonucleotides, polynucleotides, nucleic acids, or any combination thereof.ii The method of any of claims 1 to 11 wherein all of the one or more contaminant species which are present in the sample are absent from the collected apatite matrix flow-through fractions.

11 The method of any of claims I to 11 further comprising combining two or more apatite matrix flow4hrough fractions to form a purified hiomolecule 1 0 solution. whcrein the biomolecule of interest in the purified biomolecule solution comprises 90% or more of all biomolecule species in the solution, or comprises 95% or more of all biomolecule species in the solution.14. The method of claim 12 or claim 13 wherein the collected apatite matrix flow- 1 5 through fractions or the purified biomolecule solution has a UV-visible spectrophotometric absorption ratio (280nm/260nm) of 1-5 or more, preferably 1.7 or more.Ii The method of any of claims 1 to 14 wherein the hiomolecule of interest in the collected apatite matrix flow-through fractions or in the purified hiomolecule solution is present at a concentration of 0.4 -0.7mg/mI.16. The method of any of claims 1 to 15, wherein the apatite matrix is in the form of a column having an internal diameter to height ratio of 1 or more.iT The method of claim 16 wherein the apatite matrix is in the form of a column having an internal diameter of approx. 5cm and a height of approx. 7cm.18. The method of any of claims I to I? wherein the apatite matrix is in the form of a column and wherein the volume of the buffered sample does not exceed the packed volume of the apatite matrix.19. The method of any of claims 1 to 18 wherein the apatite matrix is a hydroxyapatite rnaix, preferably wherein the hydroxyapatite matrix comprises porous spheres of non-derivatized ceramic hydroxyapatite of diameter approx. 4ORm having pores of diameter of approx 600-800 A. 2(1 The use of an apatite matrix in a method defined in accordance with any of claims I to 18, preferably wherein the apatite niatrix is a hydioxyapatite matrix. more preferably wherein the hydroxyapatite comprises porous spheres of non-derivatized ceramic hydroxyapatite of diameter approx. 4ojim having 1 0 pores of diameter of approx. 600-800 A. 21. A method for producing a purified biornolecule solution from a contaminated biomolecule solution by separating a biomolecule of interest contained in the contaminated biornolecule solution from one or more contaminant species 1 5 contained in the contaminated biomolecule solution, the method comprising: a) perfoi-rning a first separation step using an iou exchange matrix, the first step comprising: i. equilibrating an ion exchange matrix with an ion exchange mati-i x equilibration solution; ii. loading the equilibrated ion exchange matrix with a volume of the contaminated hiomolecule solutioui; iii. contacting the loaded ion exchange matrix with a volume of an ion exchaiigc matrix wash solution; iv. allowing the loaded contaminated biomolecuic solution and the ion exchange matrix wash solution to flow through and exit the ion exchange matrix to form an ion exchange matrix flow-through; v. collecting fractions of the ion exchange matrix flow-through which contain the biomolecule of interest and combining the collected frachons to form an intermediate biomolecule solutioii; and b) performing a second separation step using an apatite matrix, the second step comprising: i. loading an apatite matrix with a volume of the intermediate hiornolecule solution, wherein the apatite matrix has been equilibrated with an apatite matrix equilibration solution; ii. contacting the loaded apatite matrix with a volume of an apatite matrix wash solution; iii. allowing the intermediate hiomolecule solution and the apatite matrix wash solution to flow through and exit the apatite 1 0 matrix to form an apatite matrix tiow-through; iv. coilccting fractions of the apatite matrix flow-through which contain the biomolecule of interest and which arc free of the one or more contaminant species; v. combining the collected apatite manix flow-through fractions 1 5 to form the pm-ified biomolecule solution; and wherein: 1) upon contact with and Uansit through the equilibrated ion exchange matrix and the equilibrated apatite matrix the biomolecule of interest and the one or more contaminant species do not become adsorbed to the matrices; 2) no desorption step is performed during either the first or second separation steps; 3) during the second separation step the biomolecule of interest is separated from the one or more contaminant species following differential rates of transit of the hiomolecule of interest and the one or more contaminant species through the apatite matnx.21 The method of claim 21 wherein the contaminated hiomolecule solution is produced by the steps compri sing: a) obtaining a volume of a crude preparation containing the hiomolecule of interest, optionally wherein the crude preparation is a cell-free culture supernatant; b) performing a protein precipitadon step, optionally using ammoniurn sulphate, preferably wherein the volume of the crude preparation is adjusted to be 65% saturated with ammonium sulphate; c) collecting precipitated proteins, dissolving the precipitated proteins in a resuspension solution, removing any non-dissolved proteins and collecting the dissolvate; and d) performing a dialysis step upon the dissolvate using a dialysis solution to form the contaminated biornolecule solution.1 0 23. The method of claim 22 wherein the contaminated biomolecule solution, the ion exchange mathx equilibration solution, the ion exchange matrix wash solution, the intermediate biomolecule solution, the apatite matrix equilibration solution, the apatite matrix wash solution and the purified biornolecule solution are all comprised of a buffered soludon having the same 1 5 composition.24. The method of any of claims 21 to 23 wherein each buffered solution comprises potassium phosphate pH7.2 at a concentradon of between 10- 75mM, preferably 50mM potassium phosphate pH7.2; optionally wherein each buffered solution further comprises DTT at a concentration of between 0.5-5mM, preferably I mM DTT.25. The method of claim 22 or claim 23, wherein the resuspension soluhon and the dialysis solution are comprised of a buffered solution having a composition as defiiied in accordaiice with claim 24.26. The method of any of claims 21 to 25, wherein the apatite matrix is defined in accordance with any of claims 16 to 19.27. The method of claims 26 wherein the volume of the intermediate biomolecule solution is approx. 120m1 to 150rn1.28. The method of any of claims 21 to 27, wherein the ion exchange matrix comprises beads of cross-linked agarose further comprising functional groups of CH2N(CH3) and counter ions of P043, 29 A kit of parts for use in a method defined in accordance with any of claims 21 to 28, the kit comprising a receptacle containing a quantity of an apatite matrix, preferably a bydroxyapatite matrix, more preferably a hydroxyapatite matrix defined in accordance with claim 19; the kit further comprising a receptacle containing a quantity of an ion exchange matrix, preferably an ion 1 0 exchange rnathx defined in accordance with claim 28.30. The kit according to claim 29 wherein the niatrix-containing receptacles are in the form of chrornatographic colunrns suitable for use together in a biornolecule separation method.